LLM Evaluation Framework for Hardware Design Using Python-Embedded DSLs

Home for "How To Scale Your Model", a short blog-style textbook about scaling LLMs on TPUs

Veryl: A Modern Hardware Description Language

Tensor Compute Primitives: Mid-level Intermediate Representation for Machine Learning Programs

Large Context Attention

A high-performance, zero-overhead, extensible Python compiler with built-in NumPy support

Experiment of using Tangent to autodiff triton

GPU programming related news and material links

PJRT plugin for interfacing the IREE to Jax and TensorFlow.

MLIR For Beginners tutorial

MatMul Performance Benchmarks for a Single CPU Core comparing both hand engineered and codegen kernels.

https://wavespeed.ai/ Best inference performance optimization framework for HuggingFace Diffusers on NVIDIA GPUs.

Hashed Lookup Table based Matrix Multiplication (halutmatmul) - Stella Nera accelerator

commaVQ is a dataset of compressed driving video

FauxPilot - an open-source alternative to GitHub Copilot server

Unified Executors

GPUnet is a native GPU networking layer that provides a socket abstraction over Infiniband to GPU programs for NVIDIA GPUs.

Concurrent Deferred Reference Counting

LlamaIndex is the leading framework for building LLM-powered agents over your data.

Authenticated multi-version database: sparse binary merkle tree with compact partial-tree proofs

A Compiler for the Popr Language

Diffusion Bee is the easiest way to run Stable Diffusion locally on your M1 Mac. Comes with a one-click installer. No dependencies or technical knowledge needed.

Bazel build rules for GNU M4

Apple AMX Instruction Set

100 Days of RTL

Torch Frontend for IREE

A framework for Engineering Managers

A translation validation framework for MLIR

PDK for GlobalFoundries' 180nm MCU bulk process technology (GF180MCU).

OpenMMLab's next-generation platform for general 3D object detection.